Background: This study used Item Response Theory (IRT) to model the psychometric properties of a Theory of Mind (ToM) stories task. The study also aimed to determine whether the ability to understand states of false belief in others and the ability to understand another's intention to deceive are separable skills, and to establish which is more sensitive to the presence of paranoia.
Method: A large and diverse clinical and nonclinical sample differing in levels of depression and paranoid ideation performed a ToM stories task measuring false belief and deception at first and second order.
Results: A three-factor IRT model was found to best fit the data, consisting of first- and second-order deception factors and a single false-belief factor. The first-order deception and false-belief factors had good measurement properties at low trait levels, appropriate for samples with reduced ToM ability. First-order deception and false beliefs were both sensitive to paranoid ideation with IQ predicting performance on false belief items.
Conclusions: Separable abilities were found to underlie performance on verbal ToM tasks. However, paranoia was associated with impaired performance on both false belief and deception understanding with clear impairment at the simplest level of mental state attribution.